There have been concerns over the safety of autonomous vehicles but there are good reasons why they are still a safe bet on the factory floor. Tara Hamid reports.
Uber was recently forced to suspend its self-driving car trials in North America following a fatal accident in Arizona in March, where a woman was killed after being hit by an Uber self-driving sport utility vehicle.
The accident, which marked the first fatal crash involving a self- driving vehicle and a pedestrian, once again raised critical questions over the safety of autonomous vehicles and whether the technology is ready for large-scale adoption in the urban environment.
But while driverless cars still face legal and technological challenges to overcome before being safely adopted on the roads – with Uber’s CEO Dara Khosrowshahi recently predicting driverless cars to be at least a decade away – autonomous vehicles have been extensively and increasingly adopted in industrial and logistical applications for over 50 years.
Automating the movement of materials within manufacturing operations is nothing new. Indeed, automated guided vehicles (AGVs), conveyors, sorters and shuttles have been moving and handling materials, parts and products for decades, according to “Industrial mobility: How autonomous vehicles can change manufacturing,” a new study from PwC and The Manufacturing Institute
“Thanks to advancements in sensor technology, 3D camera systems, software and artificial intelligence, these machines are increasingly capable of seeing their environments and, more importantly, learning to identify what they’re seeing,” the PwC report’s co-writer, Bobby Bono wrote in PwC’s US blog.
While the early model AGVs were guided using wires and tape, over the years the technology has become more sophisticated and today automated vehicles are mainly laser navigated, sometimes termed as Laser Guided Vehicle (LGVs).
Advanced laser scanners such as those developed by SICK provide collision protection from AGVs. Prior to the advent of laser techniques, obstacles were physically detected by deformation of contact bumpers, which activated an associated mechanical switch. Thus, the approach speed limited the impact force required to activate the collision detection system before the AGV could cause harm.
By introducing a compact safety laser scanner, the AGV was immediately allowed to travel faster as the collision prevention system could detect oncoming obstacles some distance before impact. This meant the AGV could increase its hourly load capacity and still ensure safety.
Manufacturers’ Monthly got in touch with Gary Milburn, product manager-industrial safety systems at SICK Australia to discuss how adopting laser scanners has helped improve safety in AGVs and where the technology is headed in the near future.
“After the introduction of laser scanning technology in the 1990s to replace the earlier collision- based detection systems, further developments have brought about dramatic improvements in reliability,” Milburn told Manufacturers’ Monthly.
“Over the more recent years, advancements in laser scanner technology have focused on eliminating false detections, filtering out and ignoring false responses from obstructions such as dust particles, leaves or birds.”
What this means is that the AGVs will only detect real risks, such as humans crossing their path, and will therefore ignore invalid obstructions that could lower their productive working time.
“The laser scanning technology has become a lot more intelligent, allowing it to interpret the optical data it receives back from the target more intelligently and differentiating between valid and invalid obstructions to a greater extent than we could do in the past,” Milburn said.
Automation to lower worker injuries
Apart from the obvious benefits of improved productivity and reduced labour costs, adopting AGVs for material handling has another benefit in that they help eliminate accidents and personal injuries.
In material handling operations, injuries often occur due to drivers’ lack of attention, operators driving too fast, or personnel not paying attention. Obstacle detection is therefore a key to allowing AGVs to interact with personnel safely while optimising vehicle speeds.
“Incidents are inevitable when people are driving forklifts. So if you have an automated forklift transporting goods from the storage location to the dispatch location, you obviously lower your risks because first of all you are taking people out of that area – as AGVs don’t need anyone to operate them – and in addition to that, you are taking the human error out,” Milburn said.
Statistics from WorkSafe Australia show that between 2003 and 2015 there were a total of 61 forklift fatalities nationally across a wide variety of industries. Additionally, there have been a number of serious injury claims during the time frame, ranging from around 800 to more than 1,100 serious injuries per year.
While adopting more autonomous vehicles could help lower the number of such casualties, there’s still a long way to go by Australian industries to automate their processes.
The Department of Industry, Innovation and Science’s Office of the Chief Economist recently published “Industry Insights: 1/2018 flexibility and growth” where it used data gathered by AlphaBeta advisory group to examines the major economic factors shaping Australia’s industries, regions and the economy.
AlphaBeta’s study, “The automation advantage,” analysed 20 billion hours of work in Australia to understand how every Australian job is being changed by automation and concluded that Australia could seize a $2 trillion opportunity from automation by 2030.
Dr Andrew Charlton, director of AlphaBeta research firm who undertook the study, says Australian policy makers and companies must take action now to embrace technology while also taking steps to reskill Australia’s most vulnerable workers.
According to the study, currently only nine per cent of Australia’s listed companies are making sustained investments in automation, compared with more than 20 per cent in the US and 14 per cent in leading automation nations globally.
“It would be dire for Australia’s competitiveness if companies continued with a business as usual approach,” Charlton said. “Slowing down the pace of automation, rather than accelerating it may do more harm than good, depriving Australia of the resulting productivity benefits and potentially reducing the global competitiveness of local industries.”
Automation covers a broad range of technologies including advances in Artificial Intelligence (AI), robotics and the Internet of Things (IoT). While the AlphaBeta report uses a very broad definition of automation as “the process of using machines to perform tasks that would otherwise be done by humans,” the low percentage of Australian companies investing in the technologies is still indicative of the opportunities remaining to be seized.
AGVs in high-risk environments
In addition to reducing risks of human error, AGVs can operate in conditions that humans either aren’t able to operate in, or in conditions that humans don’t operate optimally in, such as in extreme heat or cold or around hazardous materials.
“By their very nature, automation systems are designed to remove humans from menial and dangerous tasks. In the mining industry especially, the nature of work is inherently dangerous, and it has long been desirable to remove people from positions where there are risks to their safety,” Sean Carter, SICK’s product manager-identification and management told Manufacturers’ Monthly.
Driverless vehicles have already been well integrated into the mining industry in Australia. Rio Tinto introduced the Autonomous Haulage System (AHS) into its operations nearly a decade ago. Now they operate a fleet of 71 AHS trucks across three of its Western Australian Pilbara iron ore mines.
Rio Tinto recently announced that since commencing operations in 2008, their AHS- fitted haul trucks have moved a billion tonnes of ore and waste materials across five sites, about a quarter of the total material moved in the area during the period, with zero injuries, highlighting the significant safety advantages of the autonomous system.
Production facilities where the goods have a high potential risk to human life are other areas where automation is helping reduce threats to human life. Countries like China have already begun adopting robots and artificial intelligence to replace the workers in dangerous environments such as ammunition factories.
A recent report by South China Morning Post claimed that nearly 25 per cent of China’s ammunition factories have had their human workers replaced with smart machines because the factories were lacking in people who actually wanted to work in such dangerous environments.
Paper and printing industries, aircraft manufacturing, diesel engines manufacturing, railways, and healthcare facilities are some other sectors where AGVs are being successfully adopted in Australia, according to John Brittain, managing director of Materials Handling, a distributor of materials handling solutions and turnkey systems across Australia for over 35 years.
“Unlike driven materials handling equipment, which rely on the alertness of the driver, non-contact collision prevention laser scanners are normally mounted on all sides of AGVs and will detect personnel and product,” Brittain said.
“Audible and visual alarm/warning signal, together with obstacle detectors, alert personnel of an approaching AGV and assure their workplace safety. The obstacle sensors also protect property and other equipment. If there are obstacles in the threatening area of the laser sensors, the AGV will reduce speed in expectation of a stop.”
Advancements in AGV technology are mainly focused on optimising the AGV flow within the factory floor with a view on maximising efficiency and safety. SICK has been working on speed control measures that ensure the AGVs slow down gradually, rather than stopping abruptly, when they encounter obstructions.
The future developments in laser sensing technologies focus on enabling the AGV scanners to create 3D maps of their entire field of view, including the boundary walls, storage racks, etcetra, to be able to safely navigate around the obstructions.
CSIRO, in its “Advanced Manufacturing— A Roadmap for unlocking future growth opportunities for Australia” report predicts that in the near future assistive robots will work collaboratively with humans and each other, with improved sensing, awareness and decision‐making capabilities that allow full autonomy and self‐learning behaviour.
According to the report, the fast-growing mobile robots segment – with autonomous self-guiding vehicles – is expected to grow at an annual rate of 70 per cent by 2020.
The report further predicts that automation adoption is expected to spread throughout the manufacturing industries as robotics converges with other technologies such as data analytics, sensors and machine learning.